Knockout of Dip2a and Dip2c genes in tumor cells
using BE4
Next, BE4 system was analyzed in murine tumor cell B16-F10. Dip2a and
Dip2c genes were each targeted with two sgRNAs (Fig. 1G, Supplementary
Table 1). Base substitution was screened by PCR amplification,
sequencing and western blotting. pDip2a-sgRNA-1 transfection results
showed Q54Z mutation with an efficiency of 22% while pDip2c-sgRNA-1
showed mutations S72F and R73Z with a total efficiency of 33%. Knockout
of Dip2a and Dip2c proteins using sgRNAs-1 are shown in Fig. 1G, H, I,
Supplementary Fig. 2S, and 3S. Similarly, pDip2a-sgRNA-2 and
pDip2c-sgRNA-2 were transfected together with BE4 plasmid. Both
pDip2a-sgRNA-2 and pDip2c-sgRNA-2 appeared to work more efficiently and
induced 40% and 43% mutations at targeted sites respectively (Fig. 1G,
H, I, Supplementary Fig. 4S, 5S). Expression of Dip2a and Dip2c genes
from WT and mutated clones were shown in Supplementary Fig. 6S.
BE4 can induce C >T substitution in mice
To explore whether BE4 system can induce site-specific base conversion
in mice, sgRNAs targeting exon 1 of Tyr locus was designed to
inactivate tyrosinase gene (Fig. 2A). Target sequences were synthesized
and cloned into pX330-EGFP to express both sgRNA and EGFP. pTyr-sgRNA
(2.35ng/μl) and BE4 plasmid (2.65ng/μl) were co-injected into nucleus of
B6/D2F1 mouse zygotes and transplanted into surrogate mothers at
two-cell stage (Fig. 2B). pTyr-sgRNA/BE4 schematic depiction was shown
in Fig. 2C. A total of 16 live pups were obtained (Fig. 2D). Mice were
genotyped using following primer (Tyr ): Forward:
5’-AGAAATTCGAGAACTAACTG-3’, Reverse: 5’-CAGTTAGTTCTCGAATTTCT-3’ (Fig.
2E, Fig. 3A, B). PCR products were purified and sequenced to verify
targeted point mutations (Fig. 3C, D, E). A total of 10 mice (62.5%)
showed point mutations with C>T and C>A base
conversion (Fig. 3E, Table 1). Mutations occurred at high efficiency at
13-15bp in front of PAM (Fig. 3C). The editing frequencies of nonsense
mutations in Tyr locus with expected amino-acid conversion
(C>T) from arginine to a stop codon (R224Z) were 56.25% (9
out of 16) (Fig. 3E, Table 1). These mutations resulted in a mosaic
pigmentation phenotype. Some C>T substitution happened at
two-cell stage after microinjection. Several founders exhibited obvious
chimeric phenomenon with a combination of non-mutant and mutant cells
and a combination of homozygous and heterozygous cells. Founders 1#,
3#, and #5 were homozygous for nonsense mutation at targeted site with
a conversion rate of 18.75%. Founders 2#, #4, #6, #11, #13, and
#15 were showed heterozygous mutation with a frequency of 37.5%.
Founders #7, #8, #9, #10, #14, and #16 were mostly wild-type
alleles with a frequency of 37.5%. Meanwhile, C>A
substitutions in founders #12 without amino acid change was observed
with a frequency of 6.25%. No indels were detected at target site (Fig.
3E). No off-target mutations were detected at potential off-target sites
(Supplementary Fig. 1S). All results suggest that BE4 system is precise
and efficient in introducing single point mutations in vivo.